KR101555996B1 - Apparatus and Method for compositing covariance matrix using frequency diversity for low-observable target detection - Google Patents

Abstract

The present invention relates to a frequency diversity technology and more specifically, to an apparatus and a method for combining a covariance matrix using frequency diversity for target detection with LPI radar, which improve signal-to-noise ratio (SNR) highly by applying a frequency diversity method as obtaining coherent gain by a colocated structure by making a transmitter part transmit a multi carrier signal and making receiving antennas have a colocated array antenna. The apparatus of the present invention includes: a receiving part which receives a reflected target signal reflected from the target; an analog filter bank part which separates the received reflected target signal by frequency component; a plurality of covariance processing parts which re-construct an output signal of the analog filter bank part with base band signals per frequency component, calculate a calculated covariance matrix as to each frequency component, generate a compensated covariance matrix by compensating a phase in order for colocated components as to the calculated covariance matrix to have an equal phase, and calculate transmission power and noise power; a covariance matrix combining part which generates an optimum covariance matrix by calculating a weighted vector by applying the compensated covariance matrix, the transmission power and the noise power to maximum ratio combining (MRC); and an incident angle estimation part which estimates an incident angle of the reflected target signal received by the receiving part by using a direction of arrival (DOA) algorithm from the optimum covariance matrix.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a covariance matrix synthesis apparatus and a method for synthesizing a covariance matrix using frequency diversity for low-

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a frequency diversity technique, and more particularly, to an orthogonal projection matrix using a vector space formed by mutually orthogonal transmission signals together with a gain obtained by forming a virtual array, The present invention relates to a covariance matrix synthesis apparatus and method using frequency diversity for low-pass search target detection that removes original signal components existing in a received signal.

The present invention also provides a method and apparatus for generating a covariance matrix applicable to existing jammer suppression schemes even in the case where a jammer signal, a noise signal, and an original signal are contained in a received signal through removal of original signal components existing in the received signal The present invention relates to a covariance matrix synthesis apparatus and method using frequency diversity.

The present invention also provides a method for estimating an angle of arrival (AOA) of a source signal using a received signal with a jammer signal suppressed by applying an optimum weight to a beamformer using the generated covariance matrix, And an apparatus and method for covariance matrix synthesis using frequency diversity for detection.

 The Hippie Tom target detection radar is a radar that detects the hippocampus target and estimates the parameters by applying the radar signal processing technique to the hippie target such as stealth. Therefore, for detection and parameter extraction of the low-fat target, a plurality of independent radar cross-sections (RCS) for the target must be acquired and effectively added to increase the overall RCS.

 The RCS of the target varies greatly depending on the polarization, the angle from the radar to the target, and the frequency of use. Therefore, it is difficult to ensure the continued success of the target, especially the detection of the low-fat trajectory and the parameter estimation, with traditional radar signal processing methods that use one carrier frequency due to the RCS of the target's varying target.

To reduce this RCS variability, various methods have been studied in which the receiver obtains multiple independent RCSs and adds them effectively. A statistical Multiple-Input Multiple-Output (MIMO) radar obtains a plurality of independent target RCS responses along the transmission path by increasing the spatial spacing between the transmitting antennas or the receiving antennas so that the viewing angles of the targets are different from each other .

In this method, since MNs are generated by M transmitters and N receivers, RCS responses for MN targets can be obtained. Thus, this method increases the detection probability with a spatial diversity gain. However, since the coherence between received signals is not maintained between the receivers because they are spaced apart from each other, coherent gain can not be obtained, and parameter estimation such as DOA is difficult.

A colocated Coherent MIMO radar with the same position of the receive antennas forms a virtual array structure of the transmit array antennas and the receive array antennas to enable high resolution parameter estimation through coherent processing, the detection performance is advantageous over the Statistical MIMO radar in low signal to noise power ratio environments.

However, the spacing distance between the transmission antennas and the reception antennas is small, so that a spatial diversity gain can not be obtained. That is, since the RCS response of the target included in the received signal is the same in all the receive array antennas, there is a disadvantage that the RCS fluctuation phenomenon of the target can not be solved.

The frequency-dependent nature of the target's RCS has been studied experimentally for some time and there has been an analysis of the frequency spacing required to obtain independent RCSs. The frequency-agile radar employing this has increased the detection rate by transmitting multi-carrier signals. However, in the known technology, the signal processing algorithm developed by applying the frequency diversity technique to the receiver having the array antenna is not disclosed for detection and parameter estimation.

1. Korean Patent Publication No. 10-2007-0016394 2. Korean Patent No. 1193446

1. Sung-Jin Lee et al., "Performance Analysis of Monostatic RCS and Bistatic RCS for Target Division", Journal of Korea Electromagnetic Engineering Society, Vol.21, No.12, December 2010, pp.1460-1466 2. Kim et al., "Steady State Kalman Filter-Based IMM Trace Filter for Multi-Target Tracking", Journal of the Korean Society for Aeronautical Space Science, 34 (8), pp.71-78

Since the target RCS (Radar Cross Section) is a function of frequency, a single covariance matrix with significantly improved SNR (Signal-to-Noise Ratio) of each component is synthesized from a multicarrier signal transmitted on a multicarrier signal and scattered on a target And an object of the present invention is to provide a covariance matrix synthesis apparatus and method using frequency diversity for the detection of a low-fat tom target.

In order to achieve the above-described object, in order to achieve the above-mentioned object, a target RCS (Radar Cross Section) is a function of frequency, and therefore a signal-to-noise ratio (SNR) of each component is calculated from a multicarrier signal transmitted in a multi- The present invention provides a covariance matrix synthesis apparatus using frequency diversity for low-pass search detection that combines a covariance matrix greatly improved.

Wherein the covariance matrix synthesizer comprises:

A receiver for receiving a reflection target signal reflected from the target;

An analog filter bank unit for separating the received reflection target signals by respective frequency components;

The output signal of the analog filter bank unit is reconstructed into baseband signals for respective frequency components, a calculation covariance matrix for each frequency component is calculated, and phase covariance is performed so that components at the same position have the same phase with respect to the calculation covariance matrix, A plurality of covariance processing units for generating a matrix and calculating transmission power and noise power;

A covariance matrix synthesis unit for generating an optimal covariance matrix by applying weighted vectors by applying the compensated covariance matrix, transmit power, and noise power to maximum ratio combining (MRC); And

And an incident angle estimator for estimating an incident angle at which the reflection target signal is incident on the receiver using a DOA algorithm from the optimal covariance matrix.

In this case, the receiver may include a plurality of array antennas, and may be operated in a mono static mode or a bistatic mode.

In addition, the plurality of array antennas may be in the form of co-located antennas that are placed at the same position.

(여기서,

는 재구성된

주파수 성분이며,

는

의 k번째 스냅샷이고, K는 전체 스냅샷 횟수이며, H는 허미션(Hermitian)연산을 나타낸다)에 의해 이루어지는 것을 특징으로 할 수 있다.Further, the calculation of the calculation covariance matrix may be performed using Equation

(here,

Is reconstructed

Frequency component,

The

K is a total number of snapshots, and H is a Hermitian operation).

Also, the DOA algorithm may be CAPON or MUSIC (Multiple SIgnal Classification).

Also, the transmission power may be an amplitude average value of the asymmetric component from the calculation covariance matrix, and the noise power may be a value obtained by subtracting the transmission power from the diagonal component value of the calculation covariance matrix.

(여기서, w_i는 가중치 벡터이고,

는 보상 공분산 행렬을 나타낸다)에 의해 산출되는 것을 특징으로 할 수 있다.Also, the optimal covariance matrix may be expressed by Equation

(Where w _i is a weight vector,

Represents a compensation covariance matrix).

(여기서, 송신 전력

=

이고,

=

이며,

이며,

는 기대값을 나타낸다. c_i는 주파수에서의 표적 RCS 응답이고,

는 c_i의 복소수이고, x는 시간함수,

는 시간 지연, H는 허미션 연산,

은 잡음을 나타낸다)에 의해 산출되는 것을 특징으로 할 수 있다.Further, the weight vector may be expressed by Equation

(Here, transmit power

=

ego,

=

Lt;

Lt;

Represents the expected value. c _i is the target RCS response at the frequency,

Is a complex number of c _i , x is a time function,

H is a time delay, H is a hermetic operation,

Is expressed by < EMI ID = 1.0 >

On the other hand, another embodiment of the present invention includes a method of receiving a signal, comprising: receiving a reflective target signal from which a receiver is reflected from a target; Separating the received reflective target signal for each frequency component; A plurality of covariance processing units reconstructing an output signal of the analog filter bank unit into frequency-band-based baseband signals; Computing a covariance matrix for each frequency component of the plurality of covariance processing units reconstructed baseband signals; Calculating a transmit power and a noise power by generating a compensated covariance matrix by phase-compensating the plurality of covariance matrixes so that components at the same position have the same phase with respect to the computed covariance matrix; Calculating a weight vector by applying the covariance matrix, the transmit power, and the noise power to a maximum ratio combining (MRC) method to generate an optimal covariance matrix; And estimating an incident angle at which the reflection target signal is incident on the reception unit using a DOA (Direction Of Arrival) algorithm from the optimal covariance matrix of the incident angle estimation unit. A covariance matrix synthesis method using a covariance matrix is provided.

According to the present invention, a plurality of independent target RCS (Radar Cross Section) can be obtained and combined effectively even in the case of a low-fat tidal target, so that a DOA (Direction of Arrival) estimation of a target is possible.

Another advantage of the present invention is that it is possible to develop a coherent low-fat target detection algorithm based on the DOA information of the accurately estimated target.

1 is a conceptual diagram showing a transmission / reception environment of a general multi-carrier DOA radar.
FIG. 2 is a block diagram of a covariance matrix synthesis apparatus 200 using frequency diversity for low-pass search target detection according to an exemplary embodiment of the present invention. Referring to FIG.
3 is a flowchart illustrating a process of estimating an incident angle by synthesizing a covariance matrix according to an embodiment of the present invention.
4 is a graph illustrating the phase of a covariance matrix with improved linearity according to an embodiment of the present invention.
5 is a graph illustrating mean square error (MSE) for comparing performance estimates using a direction estimation algorithm from a covariance matrix according to an embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

Like reference numerals are used for similar elements in describing each drawing.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. The term "and / or" includes any combination of a plurality of related listed items or any of a plurality of related listed items.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Should not.

Hereinafter, an apparatus and method for synthesizing a covariance matrix using frequency diversity for low-pass search target detection according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a conceptual diagram showing a transmission / reception environment of a general multi-carrier DOA radar. Referring to FIG. 1, the transmitter 120 may transmit a multicarrier signal with one antenna or may transmit N narrowband signals having N different transmit antennas with different center frequencies. The receiving unit 130 receives the multicarrier signal in any form.

The transmitting unit 120 and the receiving unit 130 may operate in a general monostatic mode or may be spatially separated from each other to operate in a bistatic mode. The receiving unit 130 is composed of M array antennas 130-1 to 130-M, and the distance between the array devices has a half-wave separation distance, so that the receiving antennas are not in a distant structure, Lt; / RTI >

은 수학식 1과 같이 표현 할 수 있다.In addition, the multicarrier signals received by the respective array elements of the receiver 130 may be separated from each other by an analog band-pass filter bank in the receiver. In the following, it is assumed that the time, frequency, and phase synchronization between the transmitting and receiving units are completely achieved. The ideal frequency-to-frequency spacing of the multicarrier used by the transmitter 120 can be determined by the target size and the steering direction. If the length of the target is L and the scattering points are uniformly distributed, the frequency spacing to obtain a statistically independent RCS response

Can be expressed by Equation (1).

는 수신단에서 표적을 조향하였을 때 안테나의 수직방향과 표적이 이루는 각도이고, c는 빛의 속도를 나타낸다. 이하의 설명에서는 주파수 이격이 위수학식의 조건을 만족한 것으로 가정하고 기술하며 그렇지 못한 경우의 성능 열화 부분은 별도로 설명한다.here

Is the angle between the vertical direction of the antenna and the target when steering the target at the receiver, and c is the speed of light. In the following description, it is assumed that the frequency spacing satisfies the condition of the above equation, and the deteriorated portion of the case is described separately.

The multi-carrier signals transmitted from the transmitter 120 are matched to the target 110 and are scattered and input to the receiver 130, separated by frequency components from the receiver analog filter bank, and transited to the baseband.

주파수 성분의 기저대역 신호 백터

는 다음 수학식으로 나타낼 수 있다. In the general case where the target 110 is a long distance, the scattered signal can be assumed as a plane wave. M obtained from the M array elements 130-1 to 130-M

Baseband signal vector of frequency components

Can be expressed by the following equation.

는

주파수에서의 표적 RCS 응답이며,

는

로 정의된 평균 RCS 이며,

는 M x 1행렬이다. 위의 수식에서

(표적 RCS 응답)은 확정적인 값이 아니고 랜덤한 값을 가지게 되므로 "평균적으로 얼마인 값을 갖는다.", 혹은 수학식으로

표현하게 된다. 즉,

를 명확하게 설명하기 위한 하나의 표현 방식이 된다.here

The

The target RCS response at the frequency,

The

Lt; RTI ID = 0.0 > RCS,

Is an M x 1 matrix. In the above formula

(Target RCS response) is not a deterministic value but has a random value, so it has a "value on average", or a formula

. In other words,

This is one way of expressing it clearly.

,

, H, 그리고

는 각각 멀티캐리어 송신신호의

주파수 성분에 해당하는 시간함수, 시간 지연, 허미션(Hermitian) 연산, 그리고 잡음을 나타낸다.

는 조향벡터(steering vector)이며 산란파가 수신부(130)의 배열 안테나(130-1 내지 130-M)에 대해서

의 각도로 입사되는 경우 다음 수학식으로 나타낼 수 있다.

,

, H, and

Carrier transmission signals < RTI ID = 0.0 >

A time function corresponding to the frequency component, a time delay, a Hermitian operation, and noise.

Is a steering vector, and a scattering wave is a steering vector for the array antennas 130-1 to 130-M of the receiver 130

, The following equation can be used.

와 d는

주파수에 해당하는 파장과 수신 배열 안테나 사이의 거리를 각각 나타내며 d는 그레이팅 로브(grating lobe)를 줄이기 위해

이 되도록 하였다.

은 사용하는 송신 멀티캐리어 주파수들 중에 가장 높은 캐리어 주파수에 해당하는 파장이다. 수신된 신호

의 파워는 다음 수학식과 같이 표현할 수 있다.here,

And d is

Represents the distance between the wavelength corresponding to the frequency and the receiving array antenna, and d represents the length of the grating lobe

Respectively.

Is the wavelength corresponding to the highest carrier frequency among the used transmission multicarrier frequencies. Received signal

Can be expressed by the following equation.

와

은

주파수 성분에 해당하는 송신 전력으로 각각

과

을 나타낸다. 수학식 2를 이용하면 수신 신호

에 대한 공분산 행렬은 다음 수학식과 같이 나타낼 수 있다. Where Tr represents a trace operation

Wow

silver

The transmission power corresponding to the frequency component

and

. Using equation (2)

Can be expressed by the following equation.

는 단위행렬을 나타낸다. 공분산 행렬

의 성분 값들은 일반적으로 복소수 값을 가지며 잡음 성분이 없는 경우 j행 k열 성분은 다음 수학식과 같이 나타낼 수 있다.here

Represents an identity matrix. Covariance matrix

The component values of the j-th row and the k-th column can be represented by the following mathematical expression.

는 진폭성분이며 위상 성분 중

는

이다. 각 주파수 성분별로 공분산 행렬을 계산할 수 있으므로 N 개의 공분산 행렬 집합

이 얻어진다. here,

Is the amplitude component and the phase component

The

to be. Since a covariance matrix can be calculated for each frequency component, N covariance matrix sets

.

The covariance matrices of the covariance matrices have complex numbers even when there is no noise. Therefore, the covariance matrices of N covariance matrices in the same row and column are obtained by multiplying the N covariance matrices by weights, The phase values should be made equal.

Since the phase value of each component of the covariance matrix is a function of frequency as in Equation (6), the phase values of the same row and column components must be unified using one reference frequency.

을 기준 주파수라고 하면

의 경우 진폭 성분

은 그대로 유지하고 위상 성분

에

을 곱함으로써 위상값을 통일시킬 수 있다. 즉 이 과정에 의해

은 수학식 7의 {check boldR it_{f _{i}} LEFT ( j,`k RIGHT ) 로 변환된다.

Is referred to as a reference frequency

The amplitude component

And the phase component

on

So that the phase values can be unified. In other words,

Is converted to {check bold R it {f _ {i}} LEFT (j, `k RIGHT) in Equation (7).

을 얻을 수 있다.

에 속한 공분산 행렬들은 위상이 보상되었으므로 동일 행, 열에 속한 표적 신호성분들은 위상이 동일하다. This process is applied to all the components of the N covariance matrices to obtain a phase-compensated compensated covariance matrix set

Can be obtained.

The target signal components belonging to the same row and column have the same phase.

에 속한 공분산 행렬들을 동일한 가중치로 수학식 8처럼 단순히 더해서 합성 공분산 행렬

을 합성할 수 있다. therefore

Is simply added to Equation (8) with the same weight to obtain a composite covariance matrix

Can be synthesized.

에 포함된 잡음 전력은 일반적으로 동일하다고 가정할 수 있으므로 이하 설명에서

은 i에 관계없이

로 표기한다. In the following description, this scheme is referred to as EGC (Equal Gain Combining). Each covariance matrix

It can be assumed that the noise power included in < RTI ID = 0.0 >

Regardless of i

.

에 포함된 잡음 전력은 동일하지만 서로 통계적으로 독립이다. 그러므로 수학식 8에 의해 얻어진

의 각 성분의 SNR은 집합

에 속한 임의의 공분산 행렬의 성분 보다 SNR이 개선될 수 있다.Each covariance matrix

Are the same, but statistically independent of each other. Therefore,

The SNR of each component of

SNR can be improved over the components of any covariance matrix belonging to < RTI ID = 0.0 >

은

에 속한 공분산 행렬들에 동일한 가중치를 곱한 후 더한 것으로 각 공분산 행렬들의 위상만 통일시키고 진폭 정보를 사용하지 않은 것이다. 최종으로 합성한 공분산 행렬 각 성분들의 SNR을 극대화시키기 위해서는 동일한 가중치로 더하지 말고 진폭 정보를 고려해서 주파수 다이버시티를 최대로 활용한 가중치

을 선정한 후 다음 수학식 9를 이용해서 최종 공분산 행렬

을 합성해야 한다.

silver

The covariance matrices belonging to the covariance matrices are multiplied by the same weight, and then only the phases of the covariance matrices are unified and the amplitude information is not used. In order to maximize the SNR of each component of the covariance matrices finally synthesized, do not add the same weights but consider the frequency diversity

And then the final covariance matrix < RTI ID = 0.0 >

Should be synthesized.

의 각도로 입사되는 경우 공분산 행렬은 수학식 5에 의해서 대각(diagonal) 성분은

, 비대각(off-diagonal) 값은

값을 갖는다.Generally, the scattered waves are applied to the array antennas 130-1 to 130-M of the receiver 130

The diagonal component of the covariance matrix is expressed by Equation (5)

, The off-diagonal value is

Lt; / RTI >

값이 클수록 두드러진다. 이 결과를 수학식 9를 이용해 합성된

에 적용시켜 보면 DOA 피크값을 가장 크게 하는 조건은 다음 수학식 10과 같이 된다.When the DOA (Direction Of Arrival) algorithm such as Capon is applied to this covariance matrix, the peak value

The larger the value, the more noticeable. The result is shown in Equation 9,

The condition for maximizing the DOA peak value is as shown in Equation 10 below. &Quot; (10) "

는 다음 수학식 11과 같다.When the maximum ratio combining (MRC) theory is applied, a weighting value satisfying Equation (10)

Is expressed by the following equation (11).

을 찾는다. 따라서, 이를 수학식 9에 적용시켜 최적 공분산 행렬

을 합성할 수 있고 이를 DOA 알고리즘에 적용 시킬 수 있다. 이하 설명에서는 최적 공분산 행렬

을 얻는 방식을 MRC 방법으로 명명한다.As a result, in the receiver 130,

. Therefore, by applying this to Equation (9), the optimal covariance matrix

Can be synthesized and applied to the DOA algorithm. In the following description, the optimal covariance matrix

Is named MRC method.

FIG. 2 is a block diagram of a covariance matrix synthesis apparatus 200 using frequency diversity for low-pass search target detection according to an exemplary embodiment of the present invention. Referring to FIG. 2 is a schematic block diagram of the radar covariance matrix synthesizer 200. In FIG. 2, the covariance matrix synthesis apparatus 200 includes a reception unit 130, an analog filter bank unit 210, and a plurality of covariance processing units 220- 1 to 220-N, a covariance combining unit 230, an incident angle estimating unit 240, and the like.

1) 120 transmits a multicarrier signal using one antenna or an array antenna, and the reflective target signal reflected on the target (110 in FIG. 1) is transmitted to the first through Mth array antennas 130-1 to 130-N.

The multi-carrier signals received by the respective array elements of the first to Mth array antennas 130-1 to 130-N are input to the analog filter bank unit 210 connected to the respective array elements, and then separated by respective frequency components . To this end, the analog filter bank unit 210 is composed of a plurality of analog filter banks. Therefore, if the number of elements of the array antenna is M and the number of the multicarrier used as the transmission signal is N, a total of M analog filter banks are required and each analog filter bank of the analog filter bank unit 210 is divided into N And outputs a signal.

The output signals of the analog filter bank unit 210 are converted into a baseband signal, sampled through the ADC, and then input to the first to Nth covariance processing units 220-1 to 220-N. The first to Nth covariance processors 220-1 to 220-N reconstruct baseband signals for respective frequency components, and then calculate covariance for each frequency.

는 재구성된

주파수 성분이며

의 k번째 스냅샷을

라고 할 때 공분산 행렬 계산은 다음 수학식 12를 이용해서 할 수 있다.In Equation 2,

Is reconstructed

Frequency component

The kth snapshot of

, The covariance matrix calculation can be performed using the following equation (12).

방향에서 입사되는 하나의 산란파만 있는 경우 수학식 5를 적용해 보면

주파수 성분에 대한 이론적인 공분산 행렬은 수학식 13의 형태를 갖는다. Where K is the total number of snapshots.

If there is only one scattering wave incident in the direction, if Equation 5 is applied

The theoretical covariance matrix for the frequency component has the form of (13).

주파수 성분의 공분산 행렬의 경우 각 성분의 위상값에

을 곱함으로써 주파수 성분에 무관하게 위상을 통일시키는 과정으로 계산 공분산 행렬

을 위상이 보상된 보상 공분산 행렬

로 변환하는 과정이다 (수학식 7을 참조). The first to Nth covariance processing units 220-1 to 220-N calculate the covariance matrix for each frequency, and perform phase compensation so that the components in the same position in each matrix have the same phase. In other words

In the case of the covariance matrix of frequency components,

And the phases are unified irrespective of the frequency components. The calculation covariance matrix

Phase compensated covariance matrix < RTI ID = 0.0 >

(See Equation 7).

로부터 수신신호 전력

, 잡음 전력

을 추정한다.

는 수학식 13의 형태를 가지고 있으므로 이론적으로 비대칭(off-diagonal)에 있는 성분의 진폭 성분은

이며, 대각 성분은

이다. In addition to the phase-compensated compensated covariance matrix calculation, the first to Nth covariance processing units 220-1 to 220-N calculate the calculated covariance matrix

The received signal power

, Noise power

.

Has the form of Equation (13), so that the amplitude component of the component that is theoretically off-diagonal is

And the diagonal component is

to be.

로부터 비대칭(off-diagonal) 성분의 진폭 평균값을 구해서

을 추정할 수 있고, 대각 성분값에서 이 값을 빼서

을 추정할 수 있다. Therefore, the computed covariance matrix

The average value of the off-diagonal components is obtained

, And subtracting this value from the diagonal component value

Can be estimated.

, 수신신호 전력

, 및 잡음 전력

의 추정값은 공분산 행렬 합성부(230)에 입력되어 수학식 11에서 제시한 가중치 계산 공식

에 의해 가중치 벡터

을 계산한다. The first to Nth covariance processing units 220-1 to 220-N output the covariance matrix

, Received signal power

, And noise power

Is input to the covariance matrix synthesis unit 230 and the weight calculation formula

By weight vector

.

을 이용해서 행렬 성분들의 SNR이 최대로 개선된 최적 공분산 행렬

을 계산한다. 마지막으로 합성된

가 Capon, MUSIC(MUltiple SIgnal Classification) 등 기존의 DOA 알고리즘에 적용되어 입사각을 추정하게 된다. 표적의 DOA는 공분산 행렬의 위상 선형성에 의해 추정하게 되는데 본 발명에서 표적의 RCS 응답에 따른 가중치에 의해 수신신호 공분산 행렬의 위상 선형성이 향상됨에 따라 더욱 정확한 DOA 추정이 가능해 진다. Also, the covariance matrix synthesis unit 230 generates a covariance matrix

An optimal covariance matrix with the SNR of the matrix components maximally improved

. Finally synthesized

Is applied to existing DOA algorithms such as Capon and MUSIC (MUltiple SIgnal Classification) to estimate the incident angle. The DOA of the target is estimated by the phase linearity of the covariance matrix. According to the present invention, the phase linearity of the received signal covariance matrix is improved by the weight according to the RCS response of the target, and more accurate DOA estimation becomes possible.

3 is a flowchart illustrating a process of estimating an incident angle by synthesizing a covariance matrix according to an embodiment of the present invention. Referring to FIG. 3, when the transmitter (120 in FIG. 1) transmits a target signal to the target (110 in FIG. 1), the receiver (130 in FIG. 2) receives the reflected target signal reflected from the target (steps S310 and S320 ).

When the reflective target signal is received, the analog filter bank unit 210 (Fig. 2) separates the received reflective target signal for each frequency component (step S330).

After separation by frequency components, the separated N output signals are converted into a baseband signal and then sampled. A plurality of covariance processing units 220-1 to 220-N of FIG. 2 convert the sampled frequency- Reconstructs each frequency component, and calculates a calculation covariance matrix for each frequency component of the reconstructed baseband signals (steps S340 and S350).

Thereafter, the compensation covariance matrix is generated by phase-compensating the calculated covariance matrix so that the components at the same position have the same phase (step S360). Of course, the transmission power, the noise power, and the like are calculated at the same time.

The covariance matrix synthesis unit 230 computes a weight vector by applying the compensation covariance matrix, transmit power, and noise power to a maximum ratio combining (MRC) method and calculates an optimal covariance matrix using a covariance composition formula (Steps S370 and S380).

When the optimal covariance matrix is generated, the incident angle estimator 240 estimates the incident angle at which the reflected target signal is incident on the receiver using the DOA algorithm (step S390) from the optimal covariance matrix (step S290) .

, 수학식 8 및 수학식 9에 의해 합성된 합성 공분산 행렬

및 최적 공분산 행렬

의 첫 번째 열에 있는 성분들의 위상 변화를 나타낸 그림이다. 4 is a graph illustrating the phase of a covariance matrix with improved linearity according to an embodiment of the present invention. That is, FIG. 4 shows a phase compensated covariance matrix

, A synthetic covariance matrix synthesized by Equations (8) and (9)

And an optimal covariance matrix

The phase shift of the components in the first column of Fig.

가 가장 선형성이 우수하고 EGC(Equal Gain Combining) 방법에 의한

가 다음으로 성능이 우수하다. Since the phase values have incident angle information, the higher the linearity, the better the DOA performance is. Particularly, FIG. 4 is a result of simulation using four multi-carriers, and Ideal case is a case where there is no noise. The maximum ratio combining (MRC) proposed in the present invention

(EGC) < RTI ID = 0.0 > (Equal Gain Combining) < / RTI &

Is next in performance.

5 is a graph illustrating mean square error (MSE) for comparing performance estimates using a direction estimation algorithm from a covariance matrix according to an embodiment of the present invention. Referring to FIG. 5, the simulation is performed for four carriers and eight carriers, and the transmission power is fixed to P for performance evaluation under the same conditions.

을 사용한 경우는 싱글 캐리어를 사용한 경우 보다 주파수 다이버시티 이득으로 방향 추정 성능이 향상되었다. 그러나 MRC 방식에 의해

을 사용한 경우 성능이 가장 우수한 것을 확인할 수 있다.Therefore, if the power of P is transmitted in the case of single carrier, the multicarrier transmits signals with the power of P / N. By the EGC method

The direction estimation performance was improved by the frequency diversity gain than when the single carrier was used. However,

The best performance is obtained.

110: Target
120: transmitting unit 130: receiving unit
130-1 to 130-M: Array antenna
200: Covariance matrix synthesis device
210: Analog filter bank
220-1 to 220-N: Covariance processing unit
230: Covariance matrix synthesis unit
240: Incident angle estimating unit

Claims (9)

A receiver for receiving a reflection target signal reflected from the target;
An analog filter bank unit for separating the received reflection target signals by respective frequency components;
The output signal of the analog filter bank unit is reconstructed into baseband signals for respective frequency components, a calculation covariance matrix for each frequency component is calculated, and phase covariance is performed so that components at the same position have the same phase with respect to the calculation covariance matrix, A plurality of covariance processing units for generating a matrix and calculating transmission power and noise power;
A covariance matrix synthesis unit for generating an optimal covariance matrix by applying weighted vectors by applying the compensated covariance matrix, transmit power, and noise power to maximum ratio combining (MRC); And
An incident angle estimator for estimating an incident angle at which the reflection target signal is incident on the receiver using a DOA algorithm from the optimal covariance matrix;
And a frequency diversity covariance matrix synthesis unit for detecting a low-pitched tomographic target.
The method according to claim 1,
Wherein the receiving unit comprises a plurality of array antennas and is operated in a mono static mode or a bistatic mode.
3. The method of claim 2,
Wherein the plurality of array antennas are in the form of co-located antennas placed at the same position.
The method according to claim 1,
The computation of the computational covariance matrix may be performed using Equation

(here,

Is reconstructed

Frequency component,

The

Wherein K is a total number of snapshots, and H is a Hermitian operation). ≪ Desc / Clms Page number 20 >
The method according to claim 1,
Wherein the DOA algorithm is CAPON or MUSIC (MULTI SIIGNAL CLASSIFICATION). ≪ RTI ID = 0.0 > 8. < / RTI >
The method according to claim 1,
Wherein the transmit power is an amplitude mean value of an asymmetric component from the computed covariance matrix and the noise power is a value obtained by subtracting the transmit power from a diagonal component value of the computed covariance matrix. Covariance Matrix Synthesizer.
5. The method of claim 4,
Wherein the optimal covariance matrix is expressed by equation

(Where w _i is a weight vector,

Represents a compensated covariance matrix). ≪ RTI ID = 0.0 > [10] < / RTI > 8. The method of claim 7,
The weight vector is expressed by the following equation

(Here, transmit power

=

ego,

=

Lt;

Lt;

, C _i is the target RCS response at the frequency,

Is a complex number of c _i , x is a time function,

H is a time delay, H is a hermetic operation,

And the noise is represented by a noise. The apparatus for synthesizing a covariance matrix using frequency diversity for detecting a low-pitched target.
Receiving a reflective target signal from which the receiver is reflected from the target;
Separating the received reflective target signal for each frequency component;
A plurality of covariance processing units reconstructing an output signal of the analog filter bank unit into frequency-band-based baseband signals;
Computing a covariance matrix for each frequency component of the plurality of covariance processing units reconstructed baseband signals;
Calculating a transmit power and a noise power by generating a compensated covariance matrix by phase-compensating the plurality of covariance matrixes so that components at the same position have the same phase with respect to the computed covariance matrix;
Calculating a weight vector by applying the covariance matrix, the transmit power, and the noise power to a maximum ratio combining (MRC) method to generate an optimal covariance matrix; And
Estimating an incident angle at which the reflection target signal is incident on the receiver using an algorithm of DOA (Direction Of Arrival) from the optimal covariance matrix;
And a frequency diversity covariance matrix synthesis method for detecting a low-pitched tomographic target.

Images